Like TURF
Follow TURFTravel demand forecasts laughable
The Laughable Fiction of Travel-Demand Forecasts
By James Bacon
Bacon’s Rebellion
December 20, 2013
The Virginia Department of Transportation and regional transportation planning organizations periodically make traffic forecasts for planning purposes. The idea makes sense in the abstract — estimating future volumes of traffic is needed to determine how much, and where, we should invest in new transportation infrastructure. Unfortunately, the process is flawed. Estimates have consistently overshot the mark in recent years, feeding the sense of transportation “crisis,” justifying the construction of uneconomical projects and feeding the clamor for higher taxes.
The problem is hardly unique to Virginia. The U.S. Department of Transportation has been consistently overestimating traffic volumes for years, though rarely so egregiously as in the past year. In its 2012 Conditions and Performance Report to Congress issued December 2012, USDOT projected that Vehicle Miles Traveled in the U.S. would reach 3.3 billion. Turns out the estimate was about 11% too high. Think about that: Eleven percent off in just one year!
The Frontier Group decided to compare past USDOT projections with real-world performance. The result was the graph shown here. Writes Eric Sunquist with the State Smart Transportation Initiative: “The rolled-up trend estimates show essentially the same slope year after year, indicating that agencies providing [Highway Performance Monitoring System] data generally have not updated their models and assumptions to account for current conditions, as if they expect the year to be 1980 forever.”
Projecting long-term traffic forecasts is bad enough. Acting upon those forecasts is foolhardy. There are just too many unknowns to take the forecasts seriously. Will the price of gasoline go up or down? Will Millennials continue to reside in urban centers, sticking to their buses and bicycles as they get older, or will they move to the ‘burbs and embrace the auto-centric lifestyle? What impact will smart-cities technology have on relieving traffic congestion? Which transportation mode will innovate faster and gain more transportation market share — cars, bicycles or mass transit? How will economic restructuring affect the distribution of jobs? Will new technologies enable more people to work at home? Will driverless cars make long-distance commuting less onerous and more popular? Will developers continue building green-field projects on the metropolitan periphery or will they shift to infill and re-development? We can guess the answers but we cannot know them. And we don’t have a clue how the trends might interact in unexpected ways.
There’s one other reason to regard predictions of infinite increases in traffic and congestion with suspicion. In personal correspondence, Barry Klein, president of the Houston Property Rights Association, invokes the work of transportation theorist Yacov Zahavi to suggest why the traffic modeling systems used in Houston are thoroughly inadequate: When congestion intensifies, people change their behavior. “Travel models,” Klein writes, “historically have not included a ‘feedback loop’ so the reactions by road users when confronted with congestion was not reflected.”
Congestion is a subjective experience and people have different levels of tolerance for it. … When individuals perceive themselves to have an intolerable congestion problem they usually find a way to resolve their problem. This phenomenon is unacknowledged by transportation planners.
Here are three examples of how individuals in different social roles adapt their use of the road network and allow the commute time to stay under half an hour. Workers often-times relocate (not hard for renters), adjust their work hours and even change employers when traffic becomes irritating. Employers will relocate to parts of the region that are less congested or that put them close to the workforce that they desire. Retailers play a role, because of their habit of looking for under-served pockets of consumers and then set up stores in their proximity, which incidentally reduces congestion by giving consumers shorter shopping trips.
All these factors combine to disperse traffic over the road network. They each play a role in the on-going, unplanned but never ceasing trends that mitigate congestion.
People adapt in other ways. As traffic congestion increases, people are more likely to avail themselves of alternatives to the automobile: walking, biking, riding mass transit or working at home. Developers respond to the increased demand for convenience by building housing where more transportation options are available. Employers implement Transportation Demand Management strategies. The list goes on.
Bottom line: Long-term traffic forecasts are a fiction. Rather than spending billions of dollars expanding the transportation network in anticipation of travel demand that may or may not materialize, we should focus short-term on addressing demand that demonstrably exists right now and longer-term on achieving a better balance of land uses that generates fewer and shorter trips.